alexsuse / skdata

Data sets for machine learning in Python

skdata (scikit-data)

Skdata is a library of datasets for empirical computer science. Lots of disciplines such as machine learning, natural language processing, and computer vision have data sets. This module makes the most popular and standard datasets (even the big awkward ones) easy to access from Python programs.

The project is hosted at github: http://jaberg.github.com/skdata

Installation

There are several options for installation:

• From scratch:

  • pip install --user

• From a fresh git checkout:

  • python setup.py develop

  • python setup.py install

Documentation

See http://jaberg.github.com/skdata

News

Join the mailing list: https://groups.google.com/forum/#!forum/skdata

Thanks

Github maintains an up-to-date list of direct contributors: https://github.com/jaberg/skdata/graphs

A special thanks goes to David Cox, who provided inspiration and design guidance, and generally got this project started.

This work was supported in part by the National Science Foundation (IIS-0963668).